Skip to main content
Glama
Sign In
PI Doc MCP

PI Doc MCP

by naufal-halal
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

PI Doc MCP

Ground your AI answers in live AVEVA PI System documentation.

Python 3.11+ MCP License: MIT

What it does

PI Doc MCP is a Model Context Protocol (MCP) server that gives AI assistants like Claude direct access to the official AVEVA PI System documentation. Instead of relying on training data that may be outdated or incorrect, the AI fetches real answers from live docs — eliminating hallucination on PI-specific topics.

Results are scoped strictly to docs.aveva.com/category/pi-system. Other AVEVA product families (System Platform, CONNECT, etc.) are excluded by design.

Related MCP server: Documentation Retrieval MCP Server (DOCRET)

How it works

The server proxies the publicly accessible docs-be.aveva.com API in real time. No API key, no local doc files, no indexing step — every search and page fetch goes directly to AVEVA's documentation backend and returns content that is always up to date.

Prerequisites

Installation

1. Clone the repository

git clone https://github.com/naufal-halal/pi-doc-mcp.git
cd pi-doc-mcp

2. Install dependencies

uv sync

3. Register with Claude Code

claude mcp add pi-docs --scope user -- uv run --directory /path/to/pi-doc-mcp python server.py

Replace /path/to/pi-doc-mcp with the absolute path to the cloned folder.

4. Restart Claude Code to pick up the new MCP server. You should see pi-docs listed when you run:

claude mcp list

Available Tools

Once registered, Claude has access to three tools:

Tool

Description

search_pi_docs

Search PI System docs by keyword. Optional: bundle to scope to a specific product, n_results (default 5, max 20).

get_page

Fetch the full text of a documentation page by URL. Optional: max_chars (default 4000, max 12000).

list_pi_bundles

List all 90+ PI System documentation bundles grouped by product area (PI Server, PI Web API, Interfaces, Connectors, etc.).

Usage Examples

Ask Claude questions like:

  • "What authentication methods does PI Web API support?"

  • "What are the required tag attributes for the PI RDBMS Interface?"

  • "How do I configure buffering for a PI Interface on an interface node?"

  • "Show me the AF SDK getting started guide."

  • "List all available PI System documentation bundles."

Claude will search the live docs and cite the exact page it used.

Scope

This server covers documentation bundles under docs.aveva.com/category/pi-system, including:

  • PI Server (Windows and Linux)

  • PI Web API and AF SDK

  • PI Vision, PI DataLink, PI Manual Logger

  • PI Interfaces (OPC DA/HDA, RDBMS, UFL, Modbus, DNP3, Batch, and more)

  • PI Connectors (OPC UA, MQTT, BACnet, IEC 61850, and more)

  • Adapters for Edge Data Store

  • PI Integrators, PI SQL / OLEDB, PI OPC UA Server

Use list_pi_bundles inside Claude to see the full list.

References

Disclaimer

This project is not affiliated with, endorsed by, or supported by AVEVA. It proxies AVEVA's publicly accessible documentation API for personal and developer use. Users are responsible for complying with AVEVA's terms of use.

License

MIT © Naufal — see LICENSE

Available Tools

2 tools
get_pageB

Fetch the text of a PI System documentation page by URL.

Inputs: url (required), max_chars.

search_pi_docsA

Search AVEVA PI System documentation (scoped to docs.aveva.com/category/pi-system). Returns titles, URLs, and excerpts.

Inputs: query (required), bundle, n_results.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.9/5.0

    Scored across 2 tools

    Disambiguation5/5

    The two tools have clearly distinct purposes: search returns a list of relevant pages, while get_page retrieves the full text of a specific page. There is no overlap or ambiguity.

    Naming Consistency5/5

    Both tools follow a consistent verb_noun pattern: 'search_pi_docs' and 'get_page'. The naming is predictable and clear.

    Tool Count4/5

    With only two tools, the set is minimal but appropriate for a documentation server focused on search and retrieval. It covers the core operations without being overbearing.

    Completeness4/5

    The set provides search and full-text retrieval, which are the primary operations for a documentation server. Missing features like browsing by category or listing all pages are minor gaps that do not critically hinder agents.

    Resources

    Related MCP Connectors

    Related MCP Servers

    • mcp-ragdocs

      hannesrudolph
      A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server implementation that provides tools for retrieving and processing documentation through vector search, enabling AI assistants to augment their responses with relevant documentation context
      22 npm
      263
      MIT

    • RAG Docs MCP Server

      gorizond
      A
      license
      Not graded
      quality
      B
      maintenance
      An MCP server that provides tools for retrieving and processing documentation through vector search, enabling AI assistants to augment their responses with relevant documentation context.
      18 npm
      MIT

    • api-docs-mcp

      akshaysadanand
      A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that gives AI coding assistants access to up-to-date API documentation via RAG by crawling documentation sites, indexing them into a vector store, and enabling semantic queries.
      MIT